issues
search
Tencent
/
TencentPretrain
Tencent Pre-training framework in PyTorch & Pre-trained Model Zoo
https://github.com/Tencent/TencentPretrain/wiki
Other
1.03k
stars
142
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
SMP2020-EWECT的数据集开源使用问题
#133
pxaklbe
opened
1 week ago
3
修复ELMO预训练中bilm_target.py的seg参数缺失问题
#132
liu673
opened
1 month ago
0
add suport for baichuan2_7b
#131
Yang-Yi20
opened
3 months ago
0
请问这里的中文模型支持的最大输入序列长度是512tokens吗？超过512tokens就会被截断嘛？可不可以在微调的时候扩大模型的位置编码数量？
#130
chengzi-big
opened
3 months ago
0
add support for Qwen
#129
cjw-d
opened
5 months ago
0
add config
#128
cxy-dubug
closed
5 months ago
0
Modularize convert_tencentpretrain_to_llama.py
#127
xlhuang825
opened
6 months ago
0
optimize the way of appending.
#126
Winter523
closed
6 months ago
0
Fixed some bugs regarding activation checkpoints and updated the BPE vocabulary loader
#125
wmpscc
closed
6 months ago
0
add runtime inference for mt classifier
#124
kanson1996
opened
8 months ago
0
fix bug
#123
hhou435
closed
9 months ago
0
fix lm bug
#122
yuzhangogogo
closed
9 months ago
0
fix unit test
#121
yuzhangogogo
closed
9 months ago
0
fix unit test
#120
yuzhangogogo
closed
10 months ago
0
support LLaVa
#119
JINGZIjingzi
opened
10 months ago
0
Add CLIP model and scripts
#118
ydli-ai
opened
10 months ago
0
add mt classifier by using deepspeed
#117
yuzhangogogo
closed
10 months ago
0
add mt classifier by using deepspeed
#116
yuzhangogogo
closed
10 months ago
0
add mt classifier by using deepspeed
#115
yuzhangogogo
closed
10 months ago
0
add support for pipeline parallelism
#114
hhou435
closed
9 months ago
0
add support for pipeline parallelism
#113
hhou435
closed
10 months ago
0
单机2卡预训练LLAMA-7B报错TypeError: an integer is required (got type NoneType)
#112
smallYellowCat
opened
11 months ago
1
Refactor transformer encoder
#111
hhou435
closed
11 months ago
0
add support for model parallelism
#110
karots123
closed
11 months ago
2
fix pegasusu convert
#109
JINGZIjingzi
closed
1 year ago
0
rename argument
#108
JINGZIjingzi
opened
1 year ago
0
fix s2t prepare dataset
#107
JINGZIjingzi
closed
1 year ago
0
fix s2t prepare dataset
#106
JINGZIjingzi
closed
1 year ago
0
fix s2t prepare dataset
#105
JINGZIjingzi
closed
1 year ago
0
fix finetune s2t
#104
JINGZIjingzi
closed
1 year ago
0
fix bugs in s2t
#103
JINGZIjingzi
closed
1 year ago
0
fix bugs in s2t
#102
JINGZIjingzi
closed
1 year ago
0
Fix a bug for gqa
#101
ydli-ai
closed
1 year ago
0
Rename variables
#100
hhou435
closed
1 year ago
0
Learning rate decay
#99
Eric8932
closed
1 year ago
0
【问题】deepspeed如何对不同显存大小分配数据，我有32G和16G两种大小的GPU
#98
18liumin
opened
1 year ago
1
add GQA, BLOOM, remove APEX
#97
ydli-ai
closed
1 year ago
0
LLaMA2-70B格式转换
#96
Double-bear
opened
1 year ago
0
add gqa feature
#95
Jenine-321
closed
1 year ago
0
框架支持多机多卡吗？请问下怎么启动呢？
#94
MinhuiWan
closed
1 year ago
0
size mismatch for classifier.weight: copying a param with shape torch.Size([7, 768]) from checkpoint, the shape in current model is torch.Size([2, 768]).
#93
beat4ocean
opened
1 year ago
0
KeyError: 'd'
#92
beat4ocean
opened
1 year ago
1
Update generate_seq2seq.py
#91
Eric8932
closed
1 year ago
0
Update lstm_config.json
#90
Eric8932
closed
1 year ago
0
Update large_config.json
#89
Eric8932
closed
1 year ago
0
Update base_config.json
#88
Eric8932
closed
1 year ago
0
Update lstm_config.json
#87
Eric8932
closed
1 year ago
0
Update large_config.json
#86
Eric8932
closed
1 year ago
0
Update base_config.json
#85
Eric8932
closed
1 year ago
0
Update model.py
#84
Eric8932
closed
1 year ago
0
Next